JP2010191663A - Network management system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem in a simple configuration that real-time nature of network device information is damaged when the number of client terminals to manage network devices in cooperation with a server is increased. <P>SOLUTION: A server 1 that manages the network retains the network device information. When a device status is changed by the occurrence of an error etc. in the network device, a status change notification is transmitted from the network device to the server. On receiving the status change notification, the server updates the network device information, and transmits the status change notification to a client that manages the network in cooperation with the server. In this case, in the notification processing, the mode of notification from the server to the client is changed according to a notification occurrence frequency and notification content priority from the network device. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a network management system, and more particularly to a network management system that manages network devices with a network management server (hereinafter referred to as “server”) and a plurality of clients.

  Conventionally, in a network management system, a status change notification such as error information of a network device to be managed is notified to a server and is centrally managed in the server. However, as the number of managed units increases, there has been a problem that the load on servers and networks becomes heavy.

  A method for storing network device information inside the network device (Patent Document 1) and a method for storing it between the server and the network device (Patent Document 2) when the network device information increases and the load on the network becomes heavy is known. It has been. However, these technologies correspond to processing between network devices and servers, and do not correspond to the case where the number of client terminals that manage network devices in cooperation with servers increases. Furthermore, since the network device information is temporarily held between the network device or the server and the network device, it does not reach the server immediately, so the real-time property is impaired, and the server cannot correctly determine the information.

  In addition, it is a technology that corresponds to the case where the number of client terminals that use services on the network increases, but it monitors the service performance and communication performance provided by the server, acquires information on the degree of failure and performance degradation in the server, etc. A method of switching to an alternative service with higher response (Patent Document 3) is also known. However, this technique also does not correspond to the case where the number of client terminals that manage network devices in cooperation with servers increases. Furthermore, an alternative service providing server is necessary, and when this server is used, there is a problem that time loss due to the fact that one server cannot perform centralized management is expected, and necessary information cannot be delivered to the client in a timely manner.

JP2008-198041 Japanese Patent Laid-Open No. 2001-223694 JP 2004-171495 A

  The problem to be solved is that when the number of client terminals that manage network devices in cooperation with a server increases, the real-time property of the network device information is lost with a simple configuration.

  When the number of client terminals that manage network devices in cooperation with a server increases, the present invention ensures the real-time property of network device information with a simple configuration. The main feature is to switch the transmission route of information between the server and the client based on the above.

  In the network management system of the present invention, when it is determined that the state change notification from the network device increases and the processing load on the server is high, the processing on the server side is transferred to a part of the transfer processing to the client. There is an advantage that it can be reduced. However, even if the processing load on the server is low, if the status change notification has a high priority, the server and the client can be synchronized by treating the server to send the status change notification directly to all clients. Can do.

It is a block diagram which shows an example of the network management system to which this invention is applied. It is a block diagram which shows the server as one Example of this invention. It is a figure showing the 1st flow of the information of a state change notification. It is a figure showing the 2nd flow of the information of a state change notification. It is a flowchart which shows the transmission method determination process of a state change notification. It is a block diagram in the client which has a transfer function of a state change notification. It is a figure which displays the example of a state change notification and a priority. It is a figure which displays the transmission method by the 1st combination of the generation frequency of a state change notification, and a priority. It is a figure which displays the transmission method by the 2nd combination of the generation frequency of a state change notification, and a priority.

  A server that manages the network holds network device information. When an error or the like occurs in the network device and the device state changes, a state change notification is transmitted from the network device to the server. When the server receives the state change notification, the server updates the network device information, and transmits a network device state change notification to the client managing the network in cooperation with the server as necessary. When there are a plurality of clients, it is necessary to notify all clients in order to perform the management process in the same manner.

  The present invention improves the synchronization of network device information between the network device, the server, and the client by changing the manner of notification in the notification processing from the server to the client according to the frequency of occurrence of notification from the network device and the priority of the notification content. The objective of reducing the load on the server without losing it was realized with a simple configuration.

  FIG. 1 shows an example of a network management system to which the present invention is applied. This network management system uses a protocol such as SNMP (Simple Network Management Protocol), the server 101 has a manager 111, network devices 105, 106, 107 have agents 115, 116, 117, and the manager 111 and the agents 115, 116, 117 communicate with each other. The network 104 is managed.

  The network devices 105, 106, and 107 have management information bases (MIB) 125, 126, and 127 that hold the network device states acquired by the agents 115, 116, and 117, and the server 101 also temporarily stores network device information 121 having the same contents as the management information bases 125, 126, and 127. Yes. In this way, based on the management information bases 125, 126, and 127, the server 101 and the network devices 105, 106, and 107 share the network device state.

  In order to manage the network devices, the contents of the network device information 121 on the server 101 side and the management information bases 125, 126, and 127 on the network devices 105, 106, and 107 side must be maintained in the same state. For this reason, a request for requesting information acquisition or setting is sent from the manager 111 to the agents 115, 116, and 117, and a Response is sent from the agents 115, 116, and 117 to the manager 111 as a response to transmit information. In addition, when the agents 115, 116, and 117 detect an error, Trap is spontaneously transmitted to the server 101 as a state change notification. Request-Response control is performed mainly by the server 101, and the Trap is transmitted mainly by the network devices 105, 106, and 107.

  By the way, if there are more network devices to be managed and there is a need to perform network management at multiple bases, management clients 102 and 103 are installed at each base and connected to the server 101. Build a network management system. In this case, the server 101 mainly manages the network, and the clients 102 and 103 have the managers 112 and 113 that are partial management functions and the network device information 122 and 123 that are part of the network device information 121. When the server 101 updates the network device information 121, the update information is transmitted to the clients 102 and 103 as necessary, and the network device information 122 and 123 are also updated.

  As the number of network devices increases, the amount of management information increases, the amount of information between the server and clients increases, the number of connected management clients increases, and the control of this increases the load on the server 101. . In particular, control of state change notification by Trap requires transmission of the number of connected clients from Server 101 to one Trap when the amount of network devices to manage increases and the amount of data increases. It becomes.

  Therefore, the present invention has been proposed as a method for reducing the load on the server without losing the synchronization of the network device information among the network device, the server, and the client as much as possible. In particular, the objective is to reduce the server load when the number of clients managing the network in cooperation with the server increases with a simple configuration.

  FIG. 2 shows a server 1 as an embodiment of the present invention. This block diagram shows details of the manager 111 in FIG. In order to avoid complication of the drawing, only one network device and client are shown. When a state change notification is issued from the network device 3, it is received by the device information receiving unit 11 in the server 1, and the network device information 121 (not shown in FIG. 2) is updated.

  The server load measuring unit 12 is notified of the occurrence of the state change notification from the device information receiving unit 11, measures the occurrence frequency of the state change notification per unit time, and sets this as the load state of the server 1. The server load measuring unit 12 has a built-in timer function and continues to count state change notifications per unit time (for example, cases / second). If the measured count exceeds the threshold, it is determined that the load is high. For example, assuming 10 cases / second as a threshold, it is determined that the load is high when the state change notification of 12 cases / second is received, and the low load state is determined when 7 cases / second. This threshold is determined by the processing capability such as the CPU performance of the server 1.

  The information priority determination unit 13 receives the type of state change notification from the device information reception unit 11, determines the importance of the state occurring in the network device 3 according to the content, and determines the processing priority. The client transmission method determination unit 14 determines whether or not to transfer to the client 2 based on the load state of the server 1 and the processing priority. The device information transmission unit 15 creates device information with the transmission instruction information added to the state change notification, and transmits the device information according to the client transmission method determination unit 14. Transmission is performed to all clients when there is no transfer (FIG. 3), and to only one client when transfer is present (FIG. 4).

  FIG. 3 shows the flow of state change notification, which is when the load on the server 1 is light or when a state change notification with high priority occurs. When a status change notification occurs in the network devices 3-1, 3-2, the status change notification is sent directly from the server 1 that has received the notification to all the clients 2-1, 2-2, 2-3. .

  On the other hand, FIG. 4 shows a flow of state change notification when a load on the server 1 is heavy and a state change notification with low priority occurs. When the state change notification occurs in the network devices 3-1 and 3-2, the server 1 that has received the notification gives the client 3-1 a state change notification and an instruction to transfer the state change notification to another client. Do. The client 3-1 transfers the state change notification to the other clients 2-2 and 2-3 according to the instruction of the server 1.

  FIG. 5 is a flowchart showing a determination procedure in the client transmission method determination unit 14. The load state of the server 1 is determined (step S1 in FIG. 5), and if the load is light, it is determined that information is transmitted from the server 1 to all the clients 2-1, 2-2, 2-3 (FIG. 3). (Step S2). When the load is heavy, the priority of information is determined (step S3). As a result, when the priority is high, it is determined that information is transmitted from the server 1 to all the clients 2-1, 2-2 and 2-3 (FIG. 3) (step S2). On the other hand, if the priority is low, information is notified only to the client 2-1, and it is determined to transfer to the other clients 2-2 and 2-3 (FIG. 4) (step S4). Based on this determination, the device information transmission unit 15 creates device information in which transfer information to the client determined by the client transmission method determination unit 14 is added to the state change notification, and according to the determination by the transmission method determination unit 14, Device information is transmitted to all clients 2-1, 2-2, 2-3, or only to client 2-1.

  FIG. 6 is a block diagram showing details of the client 2-1. In FIG. 6, the client 2-1 holds the notification destination address 25 of the other clients 2-2 and 2-3. The notification address 25 is registered from the server 1 so that it can cope with a change in the system configuration. The network communication control unit 21 sends a state change notification transmitted from the server 1 to the network device information management unit 22. The network information management unit 22 updates the network device information 24 (same as 122 and 123 in FIG. 1) that stores the state of the network device in accordance with the state change notification.

  Further, the network information management unit 22 delivers the state change notification to the client control unit 23 when it is necessary to notify the other clients 2-2 and 2-3 according to the transfer information added to the state change notification. . The client transmission control unit 23 performs a state change notification transfer process based on the notification destination address 25 of the other client, and notifies the network communication control unit 21 of the result. The network communication control unit 21 transfers the state change notification to all the clients 2-2 and 2-3 registered as the notification destination address 25.

  FIG. 7 displays specific examples of state change notifications by priority. For example, when classified according to the network connection status, it is determined that the network connection failure occurs as shown in FIG. 7 as high priority, and the network connection status is determined as low priority. can do. If the table is formed so that the priority is determined based on the type of state change notification, the information priority determination unit 13 can determine the priority based on this table. FIG. 8 collectively displays transmission method determinations in which the frequency and priority are divided into two stages. The determination of A and B displayed in FIG.

  As described above in detail, it is possible to reduce the processing load on the server 1 by appropriately controlling the device information transmitted to the client 2 based on the device information received by the server 1.

  In the first embodiment, the occurrence frequency and the priority are divided into two stages, but it is also conceivable to determine the transmission method in more stages with respect to the occurrence frequency and the priority. By making decisions in multiple stages, the server load can be reduced more appropriately. FIG. 9 shows an example in which the transmission method is determined by dividing the frequency and priority into three stages.

  In the first and second embodiments, one client is instructed to transfer a state change notification to another client. However, it is also possible to divide clients into groups and instruct transfer to a plurality of clients. It is done. Client grouping can be flexibly controlled based on the occurrence frequency and priority divided into multiple stages.

1 Server 2 Client 3 Network equipment
11 Device information receiver
12 Server load measurement section
13 Information priority discriminator
14 Client transmission method judgment part
15 Device information transmitter
21 Network communication control unit
22 Network Device Information Management Department
23 Client transmission controller
24 Network device information
25 Notification address
2-1 to 2-3 clients
3-1 to 3-2 Network equipment
101 servers
102,103 clients
104 network
105-107 network equipment
111-113 manager
115-117 agents
121-123 Network device information
125 to 127 Management information base

Claims (5)

Providing a plurality of clients for managing network devices in cooperation with the server under the control of the server;
When the server receives network device status information indicating the status of each device from the network device, the server directly transmits the network device status information to all clients based on the occurrence frequency and priority. A network management system for determining whether to transmit only to a client and instruct transfer to another client. In a network management system in which the server and a plurality of clients cooperate to manage network devices under the control of the server,
The server holds network device information,
A device information receiving unit that receives a state change notification from the network device and updates the network device information;
When notified from the device information receiving unit that a state change notification has occurred, a server load measuring unit that measures the frequency of occurrence of a state change notification per unit time; and
An information priority determination unit that receives the type of state change notification from the device information reception unit, determines the importance of the state occurring in the network device according to the content, and determines the processing priority;
A client transmission method determination unit that determines whether to transfer to some clients based on the occurrence frequency and the processing priority;
If there is no transfer, the device includes a device information transmission unit that transmits the state change notification to all clients, and if the transfer is present, creates device information with the transfer information added to the state change notification and transmits the device information to some of the clients. A network management system characterized by that.   The occurrence frequency and priority are each divided into two levels, high and low, and the client transmission method determination unit determines that there is no transfer when the occurrence frequency is low or when the processing priority is high, and the occurrence frequency is 3. The network management system according to claim 2, wherein it is determined that there is a transfer when the processing priority is high and the processing priority is low.   The occurrence frequency and priority are each divided into three stages of high, medium, and low, and the client transmission method determination unit determines whether the occurrence frequency is low or the process priority is high or the occurrence frequency is medium and the process It is determined that there is no transfer when the priority is high or medium, and transfer is performed when the occurrence frequency is high and the processing priority is medium or low, or when the occurrence frequency is medium and the processing priority is low. The network management system according to claim 2, wherein the network management system is determined to be present. The some clients hold the network device information and notification addresses of other clients,
A network communication control unit for receiving a state change notification transmitted from the server and transmitting to the other client according to the transfer information;
Based on the notification destination address of the other client, the state change notification is transferred to the network communication control unit and the client transmission control unit,
3. The network information management unit according to claim 2, further comprising a network information management unit that updates the network device information according to a state change notification from the network communication control unit, and delivers a state change notification to the client transmission control unit according to the transfer information. The network management system according to -4.

Images